argumentation schemes

P1: JZP0521823196c03 CUNY025B/Walton 0 521 82319 6 August 9, 2005 19:16

THREE Argumentation Schemes

Several distinct forms of argument are identified in chapter 3 that are notdeductive or inductive in nature. These arguments are inherently pre-sumptive and defeasible, and thus they are different in nature from deduc-tive and inductive arguments. Each of the forms of argument describedin this chapter is used as a presumptive argument in a dialogue that car-ries a weight of plausibility. If the respondent accepts the premises, thenthat gives him a good reason also to accept the conclusion. But it doesnot mean that the respondent should accept the conclusion uncritically.Matching each form of argument is a set of appropriate critical questionsto ask. In a given case, there may be a balance of considerations to takeinto account. There may be some arguments in favor of the conclusionand some against it. These forms of inference are called argumentationschemes, and they represent many common types of argumentation thatare familiar in everyday conversations. They need to be evaluated in acontext of dialogue. They are used to shift a burden of proof to one sideor the other in a dialogue and need to be evaluated differently at differentstages of a dialogue. Only a few of the most important and familiar ofthese common types of argument are described in chapter 3. Others, suchas argument from consequences, are described in chapter 4.

ONE Appeal to Expert Opinion

In a critical discussion, many different facts can be relevant to the dispute.For example, in a dispute on tipping, economic data on how tipping affectsthe economy or sociological data on how tipping affects job satisfactionmay be very useful information to support or refute arguments on oneor the other side. But citing such information always rests on quotingsources, such as books or articles, which is a form of argument called

84


1. Appeal to Expert Opinion 85

‘appeal to expert opinion.’ It is frequently the case in personal, social,and political deliberations that one does not know all the relevant facts,but that even so, for reasons of time, costs, or pressing circumstances, onemust make a choice between alternative courses of action. One possibilityis to delay making a decision until more information can be collected.Frequently, this decision not to act is the most prudent course of action.But it is not always so, for delaying making a decision, by not takingany action, may itself be a course of action with significant (negative)consequences. What about the option of collecting more information? Themore information one has about the alternatives, the situation, and thelikely consequences of the available courses of action, the more informedand more practically wise one’s conclusion on how to proceed is likelyto be. But instead of trying to collect more information by doing originalresearch, it might be practical to use sources.

There can be all kinds of sources of relevant information that wouldbe helpful in a deliberation. One might have access to an encyclopedia,a dictionary, a manual, a reference book, or a computer data base. Or onemay get information from another person who has the facts. For example,if one is deliberating on the best way to get to City Hall in an unfamiliarcity, it may be very helpful to ask a passer-by who may be in a position toknow this information. You can improve your chances of getting correctinformation by choosing a source you have reason to think is reliable.But to some extent, you will have to rely on presumption or trust thatyour source is knowledgeable and honest and is not misinforming you.So you may argue, to yourself or to your companion who is with you, “Itlooks as if this passer-by knows the streets, and she says that City Hallis over that way; therefore, let’s go ahead and accept the conclusion thatCity Hall is that way.” Based on that argument, you head in the directionindicated. In this kind of case, you have acted on the basis of position-to-know argumentation.

Where a is a source of information, the following argumentationscheme represents the form of position-to-know argumentation.

ARGUMENTATION SCHEME FOR ARGUMENT FROM POSITION TO KNOW

POSITION TO KNOW PREMISE: a is in a position to know whether A istrue or false.

ASSERTION PREMISE: a asserts that A is true (false).

CONCLUSION: A may plausibly be taken to be true (false).


86 Argumentation Schemes

The form of argument can be plausible, but it is also defeasible. It can becritically questioned in a dialogue by raising doubts about the truth ofeither premise or by asking whether a is an honest (trustworthy) sourceof information. The following critical questions are appropriate for use inquestioning a position-to-know argument.

1. Is a in a position to know whether A is true (false)?2. Is a an honest (trustworthy, reliable) source?3. Did a assert that A is true (false)?

The second critical question concerns the credibility of the source. Forexample, a lawyer, when cross-examining a witness in a trial, is allowed(within controlled limits) to raise questions about the character of thewitness for honesty. If a witness has been known to lie in previouscases or can be generally shown to have bad judgment or a bad char-acter for veracity, the lawyer is allowed to use or bring these sortsof concerns forward in her cross-examination. By raising such ques-tions, the attorney could call the credibility of the witness into question,thereby influencing the jury to have doubts about the reliability of histestimony.

Much the same considerations apply in arguments outside a court-room. Position-to-know reasoning is typically used in an information-seeking type of dialogue where one has to depend on a source. It is also fre-quently used in cases where having a high-quality deliberation dialoguedepends on a prior information-seeking dialogue. In many cases of thissort, the use of information is helpful, because the information-seekingdialogue contributes to the goal of the deliberation by making the delib-eration better informed. Such a deliberation is improved by additionalrelevant information because the deliberation is made more practical bybetter fitting it to the realities of a given situation.

The appeal to expert opinion, sometimes also called ‘argument fromexpert opinion’, is an important subspecies of position-to-know reason-ing. It is based on the assumption that the source is alleged to be in aposition to know about a subject because he or she has expert knowledgeof that subject. Suppose in the context of the dialogue on tipping fromchapter 1, Helen puts forward the following argument.

Dr. Phil says that tipping lowers self-esteem.

Dr. Phil is an expert psychologist, so Helen’s argument is based on anappeal to expert opinion that can be stated as follows.



PREMISE: Dr. Phil says that tipping lowers self-esteem.

PREMISE: Dr. Phil is an expert in psychology, a field that has knowl-edge about self-esteem.

CONCLUSION: Tipping lowers self-esteem.

Helen’s argument is a plausible one that rightly carries some weight tosupport her side in the dialogue on tipping. It is an argument that couldbe countered by Bob. For example, he might cite another expert in psy-chology who disagrees with what Dr. Phil says about tipping. Even so,unless Bob criticizes the argument, it does offer a reason in support ofthe conclusion. Such arguments from expert opinion are common andimportant as evidence in trials. For example, ballistics experts and DNAexperts are often used to give expert testimony as evidence in trials.

ARGUMENTATION SCHEME FOR APPEAL TO EXPERT OPINION

MAJOR PREMISE: Source E is an expert in subject domain D containingproposition A.

MINOR PREMISE: E asserts that proposition A (in domain D) is true(false).


Appeal to expert opinion should, in most typical cases, at any rate, beseen as a plausible but defeasible form of argumentation. It is rarely wiseto treat an expert as infallible, and indeed, taking that approach can bequite dangerous, for there is quite a natural tendency to respect expertsand to defer to them unquestioningly.1 For most of us, it is not easy toquestion the opinion of an expert. It tends to verge on the impolite andthus needs to be done in a diplomatic way. But it needs to be done, inmany cases, because experts are often wrong. As a practical matter, forexample, in matters of health and finance, you can do much better if youare prepared to question the advice of an expert in a critical but politemanner. Thus it is vital to see appeal to expert opinion as defeasible, asopen to critical questioning.

1 The respect we have for authority was demonstrated by some famous experiments carriedout by the psychologist Stanley Milgram. In these experiments, subjects were asked by anauthoritative-looking scientist to administer severe electric shocks to other persons, andthey often did so.



The six basic critical questions matching the appeal to expert opinionare listed below.2

1. Expertise Question. How credible is E as an expert source?2. Field Question. Is E an expert in the field that A is in?3. Opinion Question. What did E assert that implies A?4. Trustworthiness Question. Is E personally reliable as a source?5. Consistency Question. Is A consistent with what other experts assert?6. Backup Evidence Question. Is E’s assertion based on evidence?

The idea behind using critical questions to evaluate appeals to expertopinion is dialectical. The assumption is that the issue to be settled byargumentation in a dialogue hangs on a balance of considerations. Onecan critically question an appeal to expert opinion by raising doubts aboutany of the premises. To be a genuine expert in a domain of knowledge ora technical skill, an individual must have the proper credentials and arecord of experience. It is not enough, for example, that the person is apopular celebrity. With respect to the second question, one has to lookat the exact wording of what the expert said (preferably, as quoted). Withrespect to the third question, one must be careful, for example, to checkwhether the expert may be an authority in one field (such as physics),while the proposition he is pronouncing on is in another field (such asreligion). The sixth question cites the requirement that an expert shouldbe able to back up her opinion with objective evidence. The two remainingcritical questions relate to two other implicit assumptions. The fifth ques-tion relates to “maverick” opinions, especially on issues where expertsdisagree. One can pose the consistency question by comparing A withother known evidence (and, in particular, with what experts on D otherthan a say). One can pose the trustworthiness question by expressingdoubts about whether the expert is personally reliable as a source. Forexample, one might question whether the expert is biased, for example,whether he has something to gain by making the claim put forward.

Using the appeal to expert opinion as an argument should not be seenas a substitute for getting factual evidence by scientific methods of datacollection. It is a method of argument that can be abused. However, inmany cases in deliberation and in other types of dialogue, this type ofargument, despite its fallibility, can be a valuable way of collecting usefulinformation and advice to solve a problem or make a decision. An example

2 This set of critical questions is from Douglas Walton, Appeal to Expert Opinion (UniversityPark: Penn State University Press, 1997), p. 223.



would be a case where a legislative assembly is having a discussion on abill that would legalize marijuana. One of the arguments relevant to thedebate would be the consequences of legalization. A powerful argumentagainst legalization might be the slippery slope argument linking mari-juana to increased addiction to harder drugs – a highly negative outcome,perceived as very dangerous. At this point what would very likely hap-pen in the debate is that scientific experts would be brought in to testifyon scientific findings concerning the linkages between marijuana use andthe use of harder drugs. Experts might be brought in on both sides, andtheir testimony might conflict. Both sides in the debate could questionthe scientific experts, and the dialogue might be very helpful in makingvoting on the bill more informed and intelligent. Obviously, much woulddepend on how the expert testimony was presented and how it was crit-ically questioned by the participants in the legislative assembly.

Perhaps the most familiar use of expert testimony is in the law, whereit is a very important kind of evidence, for example, in criminal trials.A familiar type of expert witness is the ballistics expert or the forensicscientist who provides, for example, evidence on DNA tests of blood orhair found at the scene of a crime. Another familiar type of expert testi-mony in the criminal trial is that given by the psychiatrist or psychologistwho is brought in to make a determination of a defendant’s ‘state of mind’when a plea of insanity has been made. Frequently, such cases turn intoa “battle of the experts,” because both sides are allowed to pay a fee forexpert witnesses thought likely to support their side of a case. Such con-flicts of expert testimony remind us that arguments based on appeal toexpert opinion are inherently subjective and presumptive in nature andare rarely conclusive. Even so, they can be valuable guides to action in adeliberation or valuable sources of informed opinion in a persuasion dia-logue. Appeals to expert opinion can, in some cases, be used fallaciouslyas well, to try to intimidate or silence one’s partner in a dialogue by sayingsomething like, “Well, you’re not an expert, are you?” This tactic can bea fallacy where it is used to try to suppress or block off legitimate criticalquestioning of what the expert said.3

In summary, then, appeal to expert opinion is a defeasible type ofargument that depends on trust in the honesty and competence of the

3 As Locke (1690) put it, when an expert is “established in dignity,” any questioning ofwhat he says by a layman may be put down as “insolence” or showing insufficient respectfor authority (Locke’s Essay, as quoted by Charles Hamblin, Fallacies (London: Methuen,1970), p. 160).



source that is consulted. If you have a choice between evidence based onthe say-so of a source that is in a position to know and objective evidencebased on scientific methods of observation, inquiry, and data collection,then it is best to give more weight to the objective evidence. But if youhave to act in a deliberation, for practical reasons, and cannot delay anylonger to collect objective information, it may be wiser to go with the say-so of a qualified expert, or someone who is in a special position to know,rather than throwing darts or relying on pure guesswork. It is a matter ofstriking a balance between qualified trust, or presumption of honesty andcompetence, on the one hand, and a skeptical attitude of doubt and criticalquestioning, on the other hand. The right balance should be decided byweighing each case individually, basing your judgment on balancing goalssuch as safety against the need to take positive action.

EXERCISE3.1

Analyze the following arguments by identifying the argumentationscheme involved. Identify the premises and conclusion of the argument.If there are any questionable aspects of the argument that should be con-sidered, identify critical questions that should be asked.

(a) Bob is lost in the jungle, in a country he is not familiar with. How-ever, he knows that Tarzan is very familiar with the terrain in thisarea, and he asks Tarzan which is the best way to get to a mountainhe wants to visit. Tarzan replies: “Don’t go across that river. It is fullof hungry crocodiles and dangerous hippos.”

(b) You go to a new dentist to have your teeth cleaned, and he recom-mends that you have a root canal and to also consider having bracesput on your teeth. He proposes to do both jobs immediately. Thisrequest seems odd to you, because you have had no problems withyour teeth recently,

(c) Herman and Louise are Canadian tourists in Australia. As he isabout to step off the curb, Herman asks: “Who has the right of way,the pedestrians or the motorists?” Louise replies, “I notice that allthe other pedestrians are giving way to the cars.”

(d) Gilbert and Joanne are having a critical discussion on the issueof whether HIV-infected surgeons ought to be allowed to operate.Joanne argues: “Doctor Dave says they ought to be allowed to, andsince this issue is a medical question, I think he should have thelast word! He is an expert.”

(e) In the dialogue on genetically modified foods (chapter 1, section 8)Sarah cited the Prince of Wales as a source to support her pointof view in the dialogue. She put forward the following argument:


2. Argument from Popular Opinion 91

“Prince Charles said that we need to rediscover a reverence for thenatural world, and that science, which lacks a spiritual dimension,should not be used to change nature.” She said, “He knows all aboutorganic agriculture. He even has his own organic garden in High-grove.”

(f) Dr. Zorba, a cancer specialist, is testifying in court in the case of aman who was bruised by his seat belt when he was rear-ended byanother car. The man later contracted testicular cancer. Dr. Zorbatestified that, in his opinion, the bruise from the seat belt was acausal factor in the development of the man’s testicular cancer. Thephysician for the insurance company testified that there is no estab-lished medical evidence that bruises or trauma caused by seat beltrestraints cause cancer.

TWO Argument from Popular Opinion

The argument from popular opinion, or appeal to popular opinion, as itis commonly called, has the following form. If a large majority (everyone,nearly everyone, etc.) accepts A as true, as shown by a poll, say, this wouldbe evidence that A is generally accepted. Or if A is common knowledge,meaning that it is an assumption that would not normally be disputed,then that is evidence that A is generally accepted. For example, in thecontest of the dialogue on tipping, neither party would dispute the state-ment that the sky is blue or the statement that people often eat food inrestaurants. If a statement is generally accepted, then that can be used asa plausible argument in favor of A.

ARGUMENTATION SCHEME FOR APPEAL TO POPULAR OPINION

GENERAL ACCEPTANCE PREMISE: A is generally accepted as true.

PRESUMPTION PREMISE: If A is generally accepted as true, that gives areason in favor of A.

CONCLUSION: There is a reason in favor of A.

By itself, this type of argument is not very strong, as it is easily shown thatthe majority is often wrong. Still, it can make a claim plausible by givinga reason in favor of it in a dialogue in which there can be reasons both forand against it. Generally, the argument from popular opinion is weakerthan the argument from expert opinion (though experts are often wrong, aswell). Argument from popular opinion is best seen as expressing a limit



on disputativeness. When you are arguing about a controversial issue,and some other proposition is relatively uncontroversial (because it isaccepted by nearly everyone or by everyone that is party to the dialogue),then that proposition can be ‘taken for granted’, or accepted tentativelyas uncontroversial. Such fine points of agreement help the dialogue goforward.

The following two critical questions match the argumentation schemefor appeal to popular opinion.

1. What evidence, such as a poll or an appeal to common knowledge,supports the claim that A is generally accepted as true?

2. Even if A is generally accepted as true, are there any good reasons fordoubting it is true?

Although arguments from popular opinion are not very strong in them-selves, they are frequently made stronger by being combined with ‘posi-tion to know’ arguments. The following case is an instance of argumentfrom popular opinion, but its strength is reinforced by an implicit argu-ment from position to know. Here is an example.

PREMISE: It is generally accepted by those who live in Cedar Rapidsthat the lake is a good place to swim in the summer.

CONCLUSION: The lake in Cedar Rapids is (plausibly) a good place toswim in the summer.

The implicit assumption that makes this appeal to popular opinion plausi-ble is that since the people who live in Cedar Rapids are normally familiarwith the area, they may be assumed to be in a position to know whether aparticular lake in the area is a good place for swimming in the summer ornot. Therefore, if the people who live in Cedar Rapids think that the lakeis a good place to swim in the summer, it is a plausible and reasonablysafe assumption (in the absence of any evidence to the contrary) that thelake is a good place to swim in the summer. This additional factor formsa chain of argumentation by joining argument from position to know tothe appeal to popular opinion. Such a joining is called a bolstering of theappeal to popular opinion, meaning that the conjoined argument increasesthe plausibility of the appeal to popular opinion. An instance of bolster-ing can be identified by stating the implicit premise, as in the examplebelow, relating to the argument just above.



IMPLICIT PREMISE: The people who live in Cedar Rapids are in a posi-tion to know whether the lake in Cedar Rapids is a good place to swimin the summer.

Of course, they could be wrong, because of some recent contamination ofthe lake that nobody knows about, for example. But in the absence of anycountervailing developments of this sort, the conclusion that the lake isa good place to swim is a plausible presumption. It is often important toidentify the implicit premise that bolsters the plausibility of the appealto popular opinion, in order to be aware of what makes the argumentplausible.

A practical form of the argument from popular opinion is called theargument from popular practice. This form of argument is connected to‘position to know’ argumentation, because familiarity with a practice is abasis for being in a position to know whether it is generally acceptable ornot.

ARGUMENTATION SCHEME FOR ARGUMENT FROM POPULAR PRACTICE

A is a popular practice among those who are familiar with what isacceptable or not with regard to A.

If A is a popular practice among those familiar with what is acceptableor not with regard to A, that gives a reason to think that A is acceptable.

Therefore, A is acceptable in this case.

An example of this type of argumentation is the following case, in whicha husband and wife are visiting Holland for the first time and have rentedbikes. They have started cycling along on a bicycle path in Holland. Heis riding behind her, thinking that riding side-by-side is not allowed. Shecalls back to him, “Ride beside me, so we can talk.” He replies, “I am notsure it is allowed.” She replies, “Everyone else is doing it.” The argumentfrom popular practice in this example dialogue is made more plausible bythe assumption that since the other couples are likely to be from the localarea, and not all tourists, they would be likely to know what is generallyaccepted in practices of riding side-by-side on the bicycle paths. So theargument from popular practice is reinforced by an implicit position toknow argument.

In still other cases, the argument from popular opinion is based noton a position to know argument but on an assumption that people havedeliberated on a particular policy or practice and have come to accept



it because they have found it a useful or good thing to do. Consider thefollowing argument concerning the Golden Rule: Do unto others as youwould they do unto you. In other words, treat others as you would like tobe treated yourself.

PREMISE: The Golden Rule is basic to every system of ethics everdevised, and everyone accepts it in some form or other.

CONCLUSION: The Golden Rule is an established moral principle thathas some weight of practical justification as a sound policy.

Here the assumption is that people have generally accepted the GoldenRule and even codified it in their systems of ethics. Such popular accep-tance lends a certain weight of presumption in favor of the Golden Ruleas an ethical principle to take seriously. It doesn’t mean that the GoldenRule can’t be questioned or criticized. It means only that the Golden Ruleshould be taken seriously in a discussion on ethical principles, becausepeople have put some thought into such matters in the past, and theirunanimity on accepting the Golden Rule indicates a presumption in itsfavor.

The two critical questions matching the argument from popular prac-tice are the following.

1. What actions or other indications show that a large majority acceptsA?

2. Even if a large majority accepts A as true, what grounds might therebe for thinking they are justified in accepting A?

With respect to the first critical question, it is frequently problematic todetermine, by asking a question, what a large majority really accepts astrue or as representing their real opinion on a matter. Public opinion pollsare often used, but much depends on how the question in the poll isworded. It may be better to go by evidence of how people act, in additionto going by what they say. But the uncertainty of verbal evidence is notthe main problem with appeals to popular opinion generally. The mainproblem resides in the asking of the second critical question.

A typical problem with appeal to popular opinion is that in manyinstances no serious attempt is made to back up the first premise, bygiving a real reason why everyone’s accepting A is a good reason why you(the respondent) should accept A. Instead, the argument is used in sucha way as to put pressure on the respondent to accept A or to feel left out



of the popular group that accepts A. For example, consider the followingargument.

You ought to buy a sports recreation vehicle, as all members of theenvironmental off-road cool people now own one of these vehicles.

Instead of giving a reason why you should buy one of these cars, thisargument tells the respondent that he will be left out of the “cool people”group if he does not buy one of these vehicles. This argument puts pressureon the respondent by appealing to his desire to be perceived as in somesocially esteemed group unless he fails to take the action advocated. Suchan argument appeals to his supposed desire to belong to some trendygroup.

So arguments from popular opinion are variable. Although they aregenerally weak arguments, in some instances they can give good reasonsto support a conclusion. In such cases, it would be unwise to ignore orreject them. But in other cases, arguments from popular opinion do notgive the required support for their conclusion and instead appeal only toa wish for belonging to an esteemed group.

EXERCISE3.2


(a) A man is steering his sailboat into an unfamiliar harbor, and he hasa choice of whether to turn left or right around a large rock. He hasobserved that all the sailboats entering the harbor before him havegone around to the left. He concludes that he should go around tothe left.

(b) If we vote to return the death penalty, we, along with a few states,will be the only jurisdictions in the Western world with a deathpenalty. Not one country in Europe has a death penalty. Canadadoesn’t have it. New Zealand doesn’t have it. Australia doesn’t haveit. It is on the books in Belgium, but there hasn’t been an executionin that country since 1945. Therefore we should not vote to returnthe death penalty.

(c) In the dialogue on genetically modified foods (chapter 1, section 8),Mark supports his point of view by arguing: “People all over theplanet have been genetically modifying animals and plants for cen-turies, nobody has worried about that, and even the scientists havelong accepted it.”



(d) Of course you should use Tartar Control toothpaste. A recent surveyshowed that 87 percent of people prefer Tartar Control toothpasteover other brands.

(e) I’m sure you will want to vote for my proposal. All highly edu-cated professionals are in favor of it and always see the merit of itimmediately.

(f) All the beautiful people who belong to the upper echelons of the richand famous have Pilotage sunglasses. So you should buy Pilotagesunglasses too.

THREE Argument from Analogy

Argument from analogy is a very commonly used kind of case-based rea-soning, where one case is held to be similar to another case in a particularrespect. Since the one case is held to have a certain property, then theother case, it is concluded, also has the same property (because the onecase is similar to the other). In outline, the argument from analogy has thefollowing form.

ARGUMENTATION SCHEME FOR ARGUMENT FROM ANALOGY

SIMILARITY PREMISE: Generally, case C1 is similar to case C2.

BASE PREMISE: A is true (false) in case C1.

CONCLUSION: A is true (false) in case C2.

This form of argument is defeasible, because any two cases will be similarto each other in certain respects, but dissimilar to each other in otherrespects. So while one case may be generally similar to another, that doesnot mean that the two cases will be similar in every respect. If they weresimilar in every respect, they would be the same case. However, two casescan be generally similar, even though there are quite important differencesbetween them.

In the dispute on tipping, Bob might use the following argument fromanalogy.

Discontinuing tipping is like taking away an animal’s source of foodby destroying its natural habitat. Taking away an animal’s source offood by destroying its natural habitat has the consequence that theanimal will painfully die by starvation and disease. So discontinu-ing tipping will take away the income of people who are struggling


3. Argument from Analogy 97

to survive in a weakening economy, with the same disastrous conse-quences.

In this argument, Bob compares two cases: the case of an animal strugglingto survive in an endangered habitat and the case of a person strugglingto survive in a weakening economy. He cites the bad consequences ofdestruction of habitat in the one case, and then postulates comparablebad consequences in the other case. Bob is using argument from conse-quences,4 but this argument is built onto an argument from analogy, basedon a comparison between the two cases. Of course, the two cases are dif-ferent in certain respects, but by comparing them, Bob puts forward aplausible argument.

There are three critical questions that are appropriate for the use ofargument from analogy.

1. Are there differences between C1 and C2 that would tend to underminethe force of the similarity cited?

2. Is A true (false) in C1?3. Is there some other case C3 that is also similar to C1, but in which A

is false (true)?

In the example above, the second critical question is easy to answer,because the base premise of the argument – that taking away an animal’shabitat has these bad consequences – is quite plausible. But asking thefirst critical question is a better avenue of criticism of the argument fromanalogy in this case. It could be argued that there is a difference betweenthe two cases. If an animal loses its habitat, there may not be any otherplace it can go or be transferred where it will have adequate supplies offood available. But in the case of a person who loses tipping income, theadditional money collected by charging the customer a higher price, oncea tip is no longer required, could be used to provide higher salaries andbenefits for the employee. At any rate, the citing of any difference of thiskind can be used to raise doubts about the argument from analogy usedin the example above.

Asking the third critical question is a very effective response in somecases, but it does not work well in all cases. The reason is that a cer-tain thoughtfulness and cleverness to devise a suitable counter-analogy is

4 This form of argument as a scheme used by Bob is to argue that an action or policy is badbecause it has bad consequences. Argument from consequences is more fully explainedin chapter 3, section 5.



required. But in some cases, use of a counter-analogy can be very effective.The following example is a classic case.

Then President Ronald Reagan, in a speech for congressional fundsto aid the Contra rebels in Nicaragua, compared the Contras to theAmerican patriots who fought in the War of Independence. A speakerin Congress opposed to sending aid to the Contras compared thesituation in Nicaragua to the war in Vietnam.

In using the argument from analogy, Reagan compared the case of theContra rebels in Nicaragua to the case of American patriots who fought inthe U.S. War of Independence. Since his audience would think the lattercase represented a highly worthwhile cause, that was good to support,and that had good consequences, their policy on this case would presum-ably be one of overwhelming support for action. By the use of argumentfrom analogy, Reagan hoped to transfer this positive attitude for supportto the present case of the Contra rebels. But his opponent in Congresswas able to counter Reagan’s argument by posing another analogy thatis also extremely powerful to the audience. Intervention in Vietnam hadextremely bad consequences for the United States and was an experiencethat nobody would care to repeat. This case stands as a powerful lessonto the effect that intervention in a foreign war can lead to a messy situ-ation that gets worse and worse, once the first steps to get involved in itare taken. Here, then, is a third case, which also appears to be similar tothe Nicaragua case in certain respects, but the outcome of interventionwas very bad.

In some cases, argument from analogy is used in an extremely aggres-sive way that packs all kinds of unstated and questionable assump-tions into the argument. The following example is taken from a letterto Chatelaine magazine, May 1982.5 Once the various implicit assump-tions in the argument are identified, it can easily be shown that they arehighly questionable and that the argument based on them is not justified.

When a murderer is found guilty, he is punished regardless of his rea-sons for killing. Similarly, anyone partaking in an abortion is guiltyof having deprived an individual of her or his right to life.

The implicit conclusion of this argument is the statement that anyonepartaking in an abortion should be punished. Why? The argument is

5 This example is from an article on arguments containing unstated assumptions: DouglasWalton and Chris Reed, “Argumentation Schemes and Enthymemes,” Synthese: An Inter-national Journal for Epistemology, Logic and Philosophy of Science, to appear.


3. Argument from Analogy 99

supported by the drawing of an analogy between two cases held to besimilar. One is the case of one person murdering another person, a crimepunishable by law. The other is the case of someone partaking in an abor-tion. The argument is that since the one type of case is similar to the other,something that is true of one should also be true of the other. The argu-ment is therefore based on the argumentation scheme for argument fromanalogy. The argument is based on the assumption that since murderersare punished, regardless of the reason for killing, by analogy, abortion par-takers should also be punished. As shown by applying the argumentationscheme for argument from analogy, this argument is based on an implicitpremise that the two kinds of cases of murder and abortion are similar.But such an assumption is highly questionable. In law, murder is a crimein which a person is killed. But in law, a fetus is not a person and has noright to life in the way a person does. Of course, this distinction is a legalone, and there is a difference between law and morality. But even so, theassumption that anyone partaking in an abortion is depriving a person ofhis or her right to life is based on the further assumption that the fetusis a person. While pro-life advocates might accept such an assumption,pro-choice advocates would reject it. They would argue that the two casesof murder and abortion are dissimilar in this respect.

In general, the first critical question for the argument from analogytends to be the most important one to focus on when evaluating argumentsfrom analogy. If one case is similar to another in a certain respect, thenthat similarity gives a certain weight of plausibility to the argument fromanalogy. But if the two cases are dissimilar in some other respect, citingthis difference tends to undermine the plausibility of the argument. Soarguments from analogy can be stronger or weaker, in different cases.

EXERCISE3.3


(a) After ingesting one milligram of substance alpha per day for ninetydays, white mice developed genetic abnormalities. Since whitemice are similar in many ways to humans, it follows that substancealpha probably produces genetic abnormalities in humans.

(b) When an individual is diagnosed as having cancer, every effortis made to kill the cancerous growth, whether by surgery, radia-tion treatment, or chemotherapy. But murderers and kidnappers arecancerous growths on society. Therefore, when these criminals are



apprehended and convicted, they should be treated like any othercancer and eliminated by capital punishment.

(c) When an elected political leader was subjected to extensive crit-icisms because of rising unemployment and a sagging economy,some argued that he should resign so that the party could select anew leader. Others said an election should be called. He replied:“People don’t change doctors just because they’re sick.”

(d) Smokers should be allowed to smoke only in private where it doesnot offend anyone else. Would any smoker walk into a restaurantand start eating half-chewed food on someone’s plate or drink aglass of water that previously held someone’s teeth? Probably not,yet they expect non-smokers to inhale smoke from the recesses oftheir lungs. My privilege and right is to choose a clean and healthylife without interference.

(e) A doctor claimed that physicians should give a medical examina-tion to every patient every year, using the following argument: “Peo-ple take their car in for servicing every few months.”

FOUR Argument from Correlation to Cause

Although scientists, particularly in the practical fields such as engineeringand medicine, sometimes make claims about causal relationships, thereis no settled scientific theory of causality (or philosophical theory, for thatmatter). It seems that the causal relationship is practical and contextualin nature. What it means to say that one state of affairs A causes anotherstate of affairs B is that A is something that can be brought about, andwhen it is brought about (or stopped), then B is also brought about (orstopped).6 At any rate, whatever causality means, the most importantkind of evidence that A causes B in any particular case is that there isa statistical correlation between A and B. For example, if a significantstatistical correlation is found between reduced incidence of heart attacksand drinking of red wine, the tentative conclusion may be drawn, as ahypothesis, that drinking red wine is the cause of the reduction in heartattacks. Recently, in fact, the causal conclusion has been drawn, based onsuch statistical findings, that drinking red wine with meals, as the Frenchdo, helps to prevent heart attacks.

As a presumptive form of reasoning, argument from correlation tocause has the following form.

6 The variables A and B stand for states of affairs that are thought of as being like propositionsbecause they can be made true (brought about) or made false by actions.


4. Argument from Correlation to Cause 101

ARGUMENTATION SCHEME FOR ARGUMENT FROM CORRELATION TO

CAUSE

CORRELATION PREMISE: There is a positive correlation between A andB.

CONCLUSION: A causes B.

A correlation is a purely statistical relationship, determined by count-ing up numbers where one event occurs in a case where another eventalso occurs. However, as noted above, causality is not a purely statisti-cal relationship (or at least, if it is, it is a very complicated kind of one),so the inference from correlation to causation cannot be evaluated on apurely statistical or numerical basis. It is best seen as a presumptive anddefeasible inference, subject to defeat as more data are collected.

One problem with arguments from correlation to cause is that theremay not be a real correlation between two events, but only seem to beone. Consider the following example:

Researchers at the Wellesley Central Hospital studied 18 patientswith rheumatoid arthritis for more than a year, testing howchanges in their symptoms were influenced by changes in theweather, but came up with no correlation between the two. “Wehypothesize that this belief results, in part at least, from peo-ple’s tendency to perceive patterns where none exist,” Dr. RonaldRedelmeier of Wellesley writes in the abstract of the study pub-lished in the latest issue of the U.S. journal Proceedings of theNational Academy of Science. . . . During his study of 18 arthri-tis patients at Wellesley, researchers gathered data on symptomstwice a month for 15 months. The patients rated the severity ofpain themselves, while doctors evaluated the degree of joint ten-derness, mobility and functioning in each patient at each assess-ment. The researchers also obtained local weather reports for sev-eral days around the time of each appointment. They interviewedthe patients about their beliefs concerning arthritis pain, and allbut two believed the effect of weather was strong. However, whenresearchers computed the correlations between pain and specificweather components mentioned by each patient, they found nopattern.7

7 Jane Gadd, “Arthritis Study Rejects Weather Link,” Globe and Mail, April 3, 1996, p. A5.



In this case there seemed to the arthritis sufferers to be a correlationbetween arthritis symptoms and weather changes. But Dr. Redelmeier’sstudy raised doubts about whether such a correlation really exists.

Another problem is that a statistical correlation between two eventscan simply be a coincidence. A sophisticated statistical study by SteffieWoolhandler and David U. Himmelstein8 citing figures from 141 countriesfound that the larger the percent of its gross national product a countryspends on weapons, the higher is its infant death rate. Woolhandler andHimmelstein concluded that there is a plausible link between militaryspending and the infant death rate: “Our findings confirm what many havesuspected – that militarism is deleterious to health even in the absenceof overt hostilities” (p. 1378). However, critics questioned whether theirfinding represents anything more than a coincidence. Dr. John Bailar, astatistician at the Harvard School of Public Health, said that the samestatistical approach could be used to show a causal link between infantmortality and the consumption of bananas.9 He questioned whether sta-tistical correlation between two things, in cases like these, is a reason toconclude that one thing causes the other.

Another critical question is whether both things correlated with eachother are really caused by some common factor that is causing both ofthem. The following case is a classic example.10

At a conference on the bond between humans and pets in Bostonin 1986, researchers reported that pets can lower blood pressurein humans, improve the survival odds of heart patients, and evenpenetrate the isolation of autistic children. According to a reportin Newsweek researchers at the conference reported on the bene-ficial effects of pet companionship. Studies showed that womenwho had owned dogs as children scored higher on self-reliance,sociability, and tolerance tests than petless women. Men who hadowned a dog “felt a greater sense of personal worth and of belong-ing and had better social skills.” Children with pets also showedgreater empathy.

In this case, there was a genuine correlation between pet ownership andhealth improvement, but both factors could well be the result of the better

8 “Militarism and Mortality,” The Lancet, June 15, 1985, pp. 1375–1378.9 “Infant Death Link Found,” Winnipeg Free Press, June 15, 1985, p. 70.

10 Douglas Walton, Informal Logic (New York: Cambridge University Press, 1989), pp. 226–227.


4. Argument from Correlation to Cause 103

than average social qualities of the people who acquire pets. This factormay lead to both pet ownership as well as better health. In a case likethis, there may be a genuine correlation between two factors A and B, butthe reason for the correlation is that some third factor C, is causing bothA and B. In such a case, it is not correct to draw the conclusion that Acauses B.

To sum up the lessons of these cases, the three main critical questionsthat should be asked, when an argument from correlation to cause is putforward, are the following.

1. Is there really a correlation between A and B?2. Is there any reason to think that the correlation is any more than a

coincidence?3. Could there be some third factor, C, that is causing both A and B?

As an example of the third critical question, consider the correlationbetween drinking red wine every day with a meal and fewer heart attacksamong men over forty. Subsequent studies showed that drinking alcoholof any sort (in moderation, meaning one or two drinks per day) was associ-ated with significantly fewer attacks within this group. The latest findingsuggested that it was the alcohol in the red wine that caused the outcomeand that drinking beer or any kind of alcohol would have the same effect.

In short, argument from correlation to cause is a legitimate and correcttype of inference of a presumptive and defeasible type, and it is extremelyuseful for practical purposes in guiding action in practical matters. Butin many cases, there is a natural human tendency to leap too quickly toa causal conclusion once a correlation has apparently been observed. Insuch cases, it is better to ask appropriate critical questions before placingtoo much weight on an argument from correlation to cause.

Here is a final word of warning. All arguments based on the statisticalclaim of a correlation should be questioned regarding how the terms weredefined in the survey. The red wine theory of heart disease preventionwas recently questioned by a group of cardiologists who pointed out thatwhile most countries require a specific cause of death to be stated, inFrance, many fatalities caused by cardiac arrest are officially put down to“sudden death.”11 This way of reporting medical statistics would meanthat the findings reported would lessen the number of heart attacks.

11 Bernard D. Kaplan, “The Attack on Red Wine’s Hearty Reputation,” Globe and Mail,September 16, 1994, p. A9.



EXERCISE3.4


(a) A report published in the Journal of the American Medical Asso-ciation found a statistical relationship between typical “male pat-tern” baldness (spreading outward from the crown of the head) andheart attacks. Men with this type of baldness were found to be 30 to300 percent more likely to suffer a heart attack than men with littleor no hair loss. Researchers hesitated to draw the conclusion thatbaldness causes heart attacks, and some speculated whether stressor a common hormonal factor might be involved (David Gelman,Carolyn Friday and Shawn D. Lewis, “A Really Bad Hair Day,”Newsweek, March 8, 1993, p. 62).

(b) A professor of medicine at the University of Toronto told a commit-tee studying welfare programs in Ontario that children from poorfamilies are twice as likely as children from more affluent familiesto die in infancy, or in accidents, and two and a half times as likelyto die from infectious diseases. He concluded that poverty is aninvisible killer that is more deadly than cancer (The Winnipeg FreePress, February 8, 1987, p. 10).

FIVE Argument from Consequences and Slippery Slope

One very common form of argumentation is used where one party in adialogue says to the other, “This action would not be good, because itcould have bad consequences.” For example, suppose you are thinking oftaking a certain medication and your doctor says, “You have high bloodpressure, and taking this medication raises blood pressure, so in your casethere would be a bad side effect of taking it.” This form of argumentation iscalled argumentum ad consequentiam, or argument from consequences(literally, it means ‘argument to consequence’). As a form of argumen-tation, it cites allegedly foreseeable consequences of a proposed actionas the premise, and the conclusion is then inferred that this course ofaction is or is not recommended. This form of reasoning can be used ina positive or negative way, as an argument to respond to a proposal thathas been put forward when two parties are having a dialogue on what todo. In argument from positive consequences, a policy or course of actionis supported by citing positive consequences of carrying out this policy


5. Argument from Consequences and Slippery Slope 105

or course of action. In argument from negative consequences, a policyor course of action is argued against by citing negative consequences ofcarrying it out.

Argument from consequences is often used in economic and politicaldeliberations where two parties (or groups) disagree on what is the bestcourse of action to pursue. For example, suppose that two persons, Boband Helen, disagree on whether tipping is generally a good custom or agood social policy that ought to be continued. Bob might use the followingargument.

PREMISE: If the practice of tipping were discontinued, unemploymentwould result.

PREMISE: Unemployment is a bad thing.

CONCLUSION: It would not be a good idea to discontinue the practiceof tipping.

In this instance, Bob has used an argument from negative consequences.By citing negative consequences of a certain policy or course of action,Bob has argued against this policy or course of action.

Argument from consequences can also be used in a positive form, tosupport a policy or action. For example, Helen might use the followingargument.

PREMISE: If the practice of tipping were discontinued, serviceproviders would have greater self-esteem.

PREMISE: Having greater self-esteem is a good thing.

CONCLUSION: The practice of tipping should be discontinued.

In this argument from consequences, Helen has cited positive conse-quences of a certain policy or course of action as a reason for supportingthat policy or course of action as being a good idea. As the last two exam-ples show, positive argument from consequences is often pitted againstnegative argument from consequences in argumentation.

Such conflicting arguments from consequences are common in politi-cal debates weighing up the alleged pros and cons of a course of action thatis being contemplated. For example, in March 1995 voters in the provinceof Quebec were having town hall meetings deliberating on whether to havea referendum giving them a choice to leave Canada and form a separatecountry or stay as a province in Canada. Some argued that the economic



consequences of separation from Canada would be highly negative forQuebec. Others argued that having a single French-speaking country sep-arate from English-speaking Canada would have positive consequencesfor French culture in Quebec. In cases of this kind of political delibera-tion, typically the argument is about the future outcomes or possibilities ofsome course of action that is unique, at least in many respects, so that thelikely consequences must be guessed or conjectured. The future can neverbe known with certainty, and guessing can be highly conjectural wheremany complex and changing variables of a real situation are involved.Hence argument from consequences is generally presumptive in natureas a kind of reasoning.

The argumentation scheme for argument from positive consequencesis the following.

ARGUMENTATION SCHEME FOR ARGUMENT FROM POSITIVE

CONSEQUENCES

PREMISE: If A is brought about, good consequences will plausiblyoccur.

CONCLUSION: A should be brought about.

The corresponding argumentation scheme for argument from negativeconsequences is the following.

ARGUMENTATION SCHEME FOR ARGUMENT FROM NEGATIVE

CONSEQUENCES

PREMISE: If A is brought about, bad consequences will plausiblyoccur.

CONCLUSION: A should not be brought about.

There are the same three critical questions matching either of these argu-mentation schemes, whether it is the positive or negative variant.

1. How strong is the probability or plausibility that these cited conse-quences will (may, might, must) occur?

2. What evidence, if any, supported the claim that these consequenceswill (may, might, must) occur if A is brought about?

3. Are there consequences of the opposite value that ought to be takeninto account?



Failure to answer any of these critical questions adequately, when askedby a dialogue partner or critic, casts an argument from consequences intodoubt.

A slippery slope argument is a species of negative reasoning fromconsequences, used where two parties are deliberating together and onewarns the other not to take a contemplated action, because it is a firststep in a sequence of events that will lead to some horrible outcome.What is distinctive about the slippery slope argument as a special subtypeof argument from consequences is that there is said to be a connectedsequence of actions, such that once the first action in the series is carriedout, a sequence of other actions will follow, so that once the sequencestarts there is no stopping it, until (eventually) the horrible outcome comesabout. This particularly horrible outcome is the final event in the sequenceand represents something that would very definitely go against goals thatare important for the participant in the deliberation who is being warned,for example, it might be his personal safety or security.

There are several types of slippery slope arguments, but the generalform of the most common type of the slippery slope argument can becharacterized by the following argumentation scheme. A slippery slopeargument always has this recursive feature, meaning that it applies overand over again in a repeating process. This feature is defined in the recur-sive premise below.

ARGUMENTATION SCHEME FOR THE SLIPPERY SLOPE ARGUMENT

FIRST STEP PREMISE: A0 is up for consideration as a proposal thatseems initially like something that should be brought about.

RECURSIVE PREMISE: Bringing up A0 would plausibly lead (in thegiven circumstances, as far as we know) to A1, which would in turnplausibly lead to A2, and so forth, through the sequence A2, . . . , An.

BAD OUTCOME PREMISE: An is a horrible (disastrous, bad) outcome.

CONCLUSION: A0 should not be brought about.

The characteristic idea of the slippery slope argument is that once you takethat first action in the sequence, it is like pushing off from the top of anOlympic ski-jump run. Once you have kicked off, turning back becomesharder and harder. At some ill-defined point or gray area, there is noturning back. Once you are into this area, there is only one way to go:faster and faster down the slope until you hit the bottom. So if you don’twant to go careening down the slope out of control and hit the bottom



(with disastrous consequences of personal injury), the message is thatyou had better not take that first step at all.

Slippery slope arguments are often used to give reason not to starttaking drugs, because once you start taking an addictive substance, itgets harder and harder to stop. At some point that cannot be definedspecifically for each individual, you are “hooked,” and the outcome isthat your life will be badly damaged or possibly even destroyed. Once aperson begins to use drugs, physical dependency can make it very difficultto stop, and the sequence of subsequent outcomes can have all kinds ofother bad consequences that eventually result in a painful life of substancedependency and eventually an unpleasant death. In this kind of case,the basis of the slippery slope is a physical addiction and dependency,caused by the body’s reaction to the substance. As you consume more ofthe substance, you need more to keep up the habit. There are also variantson this argument that cite an alleged sequence leading from one addictivesubstance to another. Here the linkages are harder to prove, and debatesabout such arguments are highly controversial. One case of this sort is theargument that the decriminalization of marijuana would be the first stepin a sequence that would lead to greater drug use and a progression to theuse of harder drugs, such as heroin, and eventually to a society where theuse of all kinds of drugs has become common and uncontrollable, with allthe social problems attending widespread drug abuse.12 A variant on thiscase with an even weaker linkage is the argument used when one personwarns another not to start smoking, arguing it may lead to the use of otheraddictive substances, like marijuana, which in turn may lead to the useof harder drugs.

Slippery slope arguments are frequently used in disputes on ethicsand public policy. In the case of Texas v. Johnson (1989) the issue con-cerned a man who burned an American flag during a political demonstra-tion in Dallas to protest policies of the Reagan administration. The caseof whether he should be convicted of “desecrating a venerated object”eventually went to the U.S. Supreme Court, where it was ruled that, inthis case, the flag-burning was an “overtly expressive” act and was thusprotected under free speech (First Amendment). Justice Brennan used aslippery slope argument as part of the rationale for his decision.

We perceive no basis on which to hold that the principle underlyingour decision in Schacht does not apply to this case. To conclude that

12 Ralph Johnson and J. Anthony Blair, Logical Self-defense, 2nd ed. (Toronto: McGraw-HillRyerson, 1983), pp. 161–162.



the Government may permit designated symbols to be used to com-municate only a limited set of messages would be to enter territoryhaving no discernible or defensible boundaries. Could the Govern-ment, on this theory, prohibit the burning of state flags? Of copies ofthe Presidential seal? Of the Constitution? In evaluating these choicesunder the First Amendment, how would we decide which symbolswere sufficiently special to warrant this unique status? To do so, wewould be forced to consult our own political preferences, and imposethem on the citizenry, in the very way that the First Amendment for-bids us to do so.13

The argument was that once the burning of the flag in a case such as theone above were prohibited by law, it would set a precedent for banningmany other kinds of acts, such as burning other objects that representedfederal or state governments. To have to enforce all these infringementswould be costly, and the value of doing so seems dubious. It would lead tothe imposition of political preferences in banning all kinds of expressiveacts that would pose conflicts with the First Amendment.

In a Doonesbury cartoon14 one character warned another of the perilsof banning “physical desecration of the flag” by citing a number of morespecific steps in the sequence of a slippery slope.

. . . “physical desecration” is a tricky business. . . . For instance, willit be illegal to burn a paper flag? Or to tear up a photo of a flag?How about cutting a cake decorated with a flag? And what about flagclothing? Are you a patriot if you wear a flag T-shirt, but a felon ifyou wear flag pants? And what does that make Uncle Sam? And whatabout art – who decides whether a flag painting is a desecration oran homage? Also, what about other national symbols, like the eagleor the Statue of Liberty? Or state flags? Or the confederate flag? Allsacred to somebody – should they be protected? Also, since burningis the only sanctioned way of disposing of a worn-out flag, aren’t wereally outlawing an idea instead of an act? And, if so, what otherideas do we outlaw?

Here we can see that what propels the sequence of reasoning down theslope is not only the idea of setting a precedent, but also the linguisticdifficulty of circumscribing a vague term such as ‘physical desecration’.Once it has been applied to one object, like a flag, it is hard to stop it fromalso being applied to other objects, like a cake or a T-shirt. This vaguenessis the gray area of the slippery slope.

13 Texas v. Johnson, 1989, 10.14 G. B. Trudeau, August 12, 1989.



Slippery slope arguments conforming to the premises and conclusionof the argumentation scheme can be reasonable as presumptive argumentsin a dialogue, provided all the connected steps in the sequences, linkingthe first step with the final (horrible) outcome, are adequately filled inand justified. In some cases, however, these links in the reasoning are notsupported adequately, and the slope argument is not plausible. In somecases, it is even used unconvincingly as a fear appeal argument. The prob-lem with fear appeal arguments is that they can easily backfire if the fearappeal is too exaggerated and unconvincing. For example, in the cultfilm Reefer Madness, teenagers were warned of the dangers of smokingmarijuana, but the evidence was dubious, according to what was knownat the time. Hence the argument was an unconvincing fear appeal argu-ment that the audiences found funny. Generally, empirical evidence isneeded to back up a slippery slope argument adequately. Since the periodwhen this film was popular, considerable empirical evidence of the nega-tive consequences of taking marijuana has been found. But how strong thelink is between taking marijuana and progressing to harder drugs remainsa subject of some controversy. Even so, drugs that are addictive are verydangerous for some people to try. So the burden of proof in such casesshould be weighted toward caution. If even trying such a drug might bedangerous, urging a person to be cautious by not taking the first step couldbe a reasonable slippery slope argument.

Generally, in evaluating slippery slope reasoning, it is best to beginby identifying the three premises of the argument. The first-step premisepostulates the first step in the sequence. The recursive premise describesthe mechanism of the slope – the repeating or propelling factor that drivesthe sequence along past a point where the sequence can be stopped. Thebad outcome premise cites the horrible outcome, supposedly the finalevent in the sequence. The most important appropriate critical questionsfor a slippery slope argument concern part 2, the sequence of steps in theargument.

1. What intervening propositions in the sequence linking up A0 with An

are actually given?2. What other steps are required to fill in the sequence of events to make

it plausible?3. What are the weakest links in the sequence, where specific critical

questions should be asked about whether one event will really leadto another?



How strong a slippery slope argument needs to be depends on how stronga claim is made in the conclusion – does it say that the horrible outcomemay, will, or must occur? The stronger the claim, the stronger the argumentneeded to back it up. However, since slippery slope arguments are aboutthe future, any of them that have a “must” in the conclusion (or anywording indicating inevitability) should be viewed very skeptically.

One must be very careful to distinguish between instances of argu-ments from negative consequences and instances of slippery slopearguments because, as indicated above, the slippery slope argumentsdescribed above are species of argumentation from negative conse-quences. What marks out the slippery slope argument as a special type ofargument from negative consequences is that the slippery slope argumentalways has the characteristic recursive premise. This premise describes asequence of actions where the argument moves forward from one step tothe next by some repeating process of mechanism that drives the actionsor consequences past the point where the sequence can be stopped.Many instances of argument from consequences do have such a recur-sive premise stated as part of the argument. If an argument from negativeconsequences does not have this recursive feature, it should not be clas-sified as a slippery slope argument.

EXERCISE3.5

Analyze the following dialogues, classifying the argument used by identi-fying its argumentation scheme. Identify the premises and conclusion ofthe argument. If there are any questionable aspects of the argument thatshould be considered, identify critical questions that should be asked.

(a) Pierre and Mary are arguing about the issue of Quebec separatingfrom the rest of Canada. Pierre maintains that separation would bea good thing, because it would preserve the Francophone culturalheritage. Mary argues that if Quebec separates, there would be mas-sive unemployment, especially in Quebec, where there are manyfederal government employees.

(b) Bob is about to try to fix his radio by picking up a live wire. Janewarns him, “I wouldn’t do that. You could get a nasty shock!”

(c) Anne and Fred are arguing about the issue of whether scien-tists should get permission to do laboratory research on humanembryos. Fred argues: “We must stop all research with humanembryos immediately, because it will lead to more and more exper-imental research on human fetuses, and eventually, the harvest ofhuman aborted fetuses will become such a resource for treatment of



diseases like Parkinson’s disease and muscular dystrophy, therewill be no turning back.”

(d) John and Louise are discussing the issue of whether physician-assisted suicide should be allowed in cases of terminal illness.Louise argues: “Once you allow it in cases of terminal illness, itwill lead to euthanasia of the disabled. That will mean that anyonewho is regarded as troublesome will be eliminated as ‘unfit’ by thestate. Ultimately, the only citizens who will be allowed to live arethose who fit the current concept of an ideal healthy person.”

(e) In a debate on whether the practice of not allowing prayers in theschools should be continued, some participants worried that differ-ent religions might want to have special prayers representing theirown religious practices and views. One participant argued that withall the different minority groups, once you accept one kind of reli-gion as legitimate, you are going to have to accept many other kindsof religious groups as having a legitimate right to have prayers orreligious services in the classroom. This participant said: “It’s aPandora’s box. You know that Satanism is a religion too!”

(f) Trevor and Mary are arguing about the issue of whether drivers’licenses should have photographs on them. Trevor argues: “It wouldbe the first step toward a police state!”

SIX Argument from Sign

In many cases of argumentation, data observed in a case are taken as asign of something that fits a familiar pattern. Drawing an inference from apatient’s symptom taken as a visible sign of a disease or some other knowncondition is a very common form of reasoning in medical diagnosis. Forexample, suppose a patient who is feeling ill has yellow skin. Such anobservation may be taken by the physician as a sign or indicator that thepatient has hepatitis. Of course, such a provisional diagnosis is only ahypothesis. The patient may have some other liver disease or may have askin disease that does not affect the liver. This form of argumentation isdefeasible at the first stages and may lead only to a plausible conclusion inthe form of a tentative hypothesis. It may lead to further observations andtests that can be carried out. The hypothesis may be tested, for example,by analyzing a sample of the patient’s blood. Argument from sign is typi-cally a defeasible argument in an investigation that leads to the collectionof further evidence. The following classic example illustrates argumentfrom sign.


6. Argument from Sign 113

Travis and Lisa are walking along a hiking trail in Jasper NationalPark and they see some imprints on the trail. Travis examinesthem closely and says he recognizes them as bear tracks, saying.“A bear has been here.” Lisa replies, “How do you know thoseimprints are bear tracks? They don’t look big enough to be beartracks.” Travis replies, “They are the tracks of a small bear. In fact,they are the tracks of a small grizzly bear, as we can see by thesevery long claw imprints.”

In this case, Travis has presented an argument to Lisa. She has expresseddoubts that the imprints they saw are bear tracks. Travis has used argu-ment from sign to give her a reason to accept the conclusion that theyare bear tracks. Argument from sign is a presumptive type of argumentbased on a premise that, generally, findings, as observed in a case, arecharacteristic of some type of object, event, or action. The other premiseis that these characteristics or signs are present in the given case. Theconclusion is that the particular event or object in question will occur orhas occurred in this particular case.

The argumentation scheme for argument from sign is the following,where A and B are taken as these two propositions.

ARGUMENTATION SCHEME FOR ARGUMENT FROM SIGN

SPECIFIC PREMISE: A (a finding) is true in this situation.

GENERAL PREMISE: B is generally indicated as true when its sign, A,is true.

CONCLUSION: B is true in this situation.

It is easy to see why argument from sign, in the scheme displayed above,is defeasible. The general premise is not an absolute universal generaliza-tion. The one proposition being true generally (but not necessarily) indi-cates that the other will be true as well. Yellow skin may be sign of liverdysfunction. But there are cases where a patient will have yellow skin butwill not have liver dysfunction. The presence of yellow skin coloration isjust a preliminary indicator or symptom. It may have evidential value. Butthe presence of other known factors in a case may rule it out, defeating theinitial inference to the conclusion that this patient has liver dysfunction.

Argument from sign proceeds from a finding of some data in the formof a sign or indicator. The sign is some sort of observed finding. A con-clusion is then drawn that the sign indicates the presence of something



else that it is connected to. The conclusion drawn can be expressed as aproposition that some event will or has taken place or that some objector factor is present. A finding is known or taken to be a sign of someproposition inferred from it for various reasons. It could be that one typeof event is usually causally connected to the other. Or it could be that afinding of the presence of some other factor, because of some identifiablecharacteristic of that factor, is made apparent in the sign. For example,the presence of dark clouds is a sign of rain because one is commonlyassociated with the other and because dark clouds can be cited as a causeof rain. For the purposes of the argumentation scheme, the two thingsthat are connected, the finding and the conclusion drawn from it, can beexpressed in the form of propositions.

In evaluating argument from sign, the following two critical questionsare appropriate.

1. What is the strength of the correlation of the sign with the eventsignified?

2. Are there other events that would more reliably account for the sign?

Quite often, argument from sign is a weak form of argument that cannotbe relied on uncritically. Even so, it is a presumptive form of argumentthat can sometimes help point an investigation or chain of reasoning to aplausible conclusion.

In some cases, arguments from sign are predictive. For example, darkclouds or high winds might be a sign of a particular kind of weather thatwill occur in the future, such as a storm. In some cases, arguments fromsign are used in a retroductive fashion, as in the bear case where the exis-tence of the bear prints are used to reason backward in time to the conclu-sion that, in the past, the bear was present at this location. The presenceof the bear is used as a basis for explaining the presence of the tracks.15

If we look at argument from sign as a form of defeasible reasoning,we can easily see its presumptive nature. The bear tracks could be takenas a sign of something else, depending on what was observed and wherethe observations were made. Perhaps someone, using some sort of gad-get, cleverly planted the bear tracks there to mislead us or give us theimpression that there are bears in this region. But since in Jasper NationalPark, normally there are bears present, we would take the presence ofthese bear tracks as a pretty good argument that there was a bear in this

15 Sometimes the kind of argumentation exhibited in the bear tracks example is called infer-ence to the best explanation. The importance of this form of argumentation in scientificdiscovery was established by the American philosopher Charles S. Peirce.


6. Argument from Sign 115

region. However, if we were to find bear tracks in a university classroom,we would be much less inclined to leap to the conclusion that there musthave been a bear present in this classroom. We would probably expectthat these bear prints were the result of some student prank or would tryto find some other explanation, because it would not be normal for a bearto be present in a university classroom.

Characteristically, argument from sign provides a basis for making aninitial intelligent guess or hypothesis, which leads to a fuller explanationonce the fuller context of a case is filled in. For example, in the followingcase16 two initial signs of footprints and tusk-shaped wounds led to anexplanation of several rhinoceros deaths.

When conservationists at South Africa’s Pilanesberg game reservediscovered a series of systematically killed rhinos, they had twoclues to the culprits: tusk-shaped wounds on the corpses andelephant footprints in the vicinity. Although an elephant doesnot normally attack a rhinoceros, the game reserve has a numberof unsupervised, adolescent males who would normally be keptin line by bulls. Without adult role models to test themselvesagainst, the animals have become juvenile delinquents.

Here the tusk-shaped wounds and the elephant footprints offered clueson a basis of argument from sign. Two signs suggested that elephants werethe killers. But further evidence led to an even more specific hypothesis.The initial hypothesis was thereby confirmed, once the fuller context oflack of supervision of the juveniles by role models was filled in.

In some cases, there is a sequence of signs, each one of which, by itself,gives only a small weight of presumption for a conclusion. But when youput the sequence together, there is an evidence-accumulating argument.This argument uses a chain of reasoning containing several argumentsfrom sign to build up to a much more plausible weight of presumption infavor of the conclusion. In the following tale from A Study in Scarlet, Dr.Watson, looking for accommodations in London, has just been introducedto Sherlock Holmes. Holmes used the following sequence of reasoning toarrive at the conclusion that Watson has just returned from Afghanistan:

Here is a gentleman of a medical type, but with the air of a militaryman. Clearly an army doctor, then. He has just come from the tropics,for his face is dark, and that is not the natural tint of his skin, forhis wrists are fair. He has undergone hardship and sickness, as his

16 Michael Kesterton, “Social Studies,” Globe and Mail, October 13, 1994, p. A20.



haggard face says clearly. His left arm has been injured. He holds it ina stiff and unnatural manner. Where in the tropics could an Englisharmy doctor have seen much hardship and got his arm wounded?Clearly in Afghanistan.17

Holmes made a guess, in this case. It was only a plausible hypothesis. Butas the individual instances of argument from sign built up evidence as hissequence of reasoning proceeded, the accumulation of evidence made hisconclusion a plausible one to draw (at least, according to the story). Thecontext of the case also played an important role here, because Holmeswas aware that there had recently been a war in Afghanistan in whichmany British men of that era had taken part. Hence the best explanationof all the phenomena that Holmes observed – the tanned skin, the injuredarm, and so forth – would be that Watson had been a participant in theAfghanistan campaign. Of course, it was only a hypothesis. But Holmes’sconclusion drawn by argument from sign was a clever bit of reasoning,characteristic of the fictional detective’s careful observations and powerof drawing inferences from them.

EXERCISE3.6


1. Barbara had a runny nose, fever, cough, and nasal congestion. On thefifth day, red spots appeared on her body. Therefore, Barbara has themeasles (or rubeola).

2. Jane had a runny nose, swollen glands, and a slight fever. On the thirdday, red spots appeared on her face and neck, and then faded after twodays. The physician suspected Jane had German measles (rubella). Hegave Jane a blood test that confirmed his diagnosis.

SEVEN Argument from Commitment

In argument from commitment, the proponent takes as premise a propo-sition that the respondent is committed to and uses it to press the

17 Arthur Conan Doyle, The Complete Sherlock Holmes, vol. 1 (New York: Doubleday, 1932),p. 11.


7. Argument from Commitment 117

respondent to concede another proposition that follows by inference fromthat premise. An example is the following case.18

Bob: Ed, you are a communist, aren’t you?

Ed: Of course. You know that.

Bob: Well, then you should be on the side of the union in this recent labordispute.

In this case, let’s say, Ed has frequently advocated communism in the pastand has often been known, for example, to shout, “Power to the people!”in demonstrations. Also, in this case, when asked, Ed admitted to Bobthat he is a communist. So the conclusion may be drawn that Ed is acommunist or that he committed to communism, as we might put it. Giventhis general commitment of Ed’s, Bob draws the plausible conclusion thatEd sides with the union in a particular labor dispute. Of course, Ed mightnot, but since a communist would normally be strongly in support of theunion side in a labor dispute, it can be inferred defeasibly that Ed is onthe union side. Of course, if the dialogue went on, and Ed declared thatin this particular case he was not with the union, Bob’s conclusion wouldhave to be withdrawn, on the balance of considerations.

Argument from commitment has the following general form, where ais a participant in a dialogue, and A and B are statements, as usual.

ARGUMENTATION SCHEME FOR ARGUMENT FROM COMMITMENT

COMMITMENT EVIDENCE PREMISE: In this case it was shown that a iscommitted to proposition A, according to the evidence of what hesaid or did.

LINKAGE OF COMMITMENTS PREMISE: Generally, when an arguer iscommitted to A, it can be inferred that he is also committed to B.

CONCLUSION: In this case, a is committed to B.

There are two critical questions that are appropriate for responding to theuse of argument from commitment.

1. What evidence in the case supports the claim that a is committed toA, and does it include contrary evidence, indicating that a might notbe committed to A?

18 Douglas Walton, Argumentation Schemes for Presumptive Reasoning (Mahwah, N.J.:Erlbaum, 1996), p. 55.



2. Is there room for questioning whether there is an exception in thiscase to the general rule that commitment to A implies commitmentto B?

In examining the second critical question, one should ask whether propo-sition B, as cited in the linkage of commitments premise, is identical to theproposition A as cited in it. If not, some discussion on what exactly is thenature of the relationship between the two propositions can be helpful.

If the respondent in the dialogue asks either critical question,where the proponent has just used argument from commitmentat the last move, then the burden of proof is shifted back to theproponent’s side. Unless the proponent can answer the criticalquestion adequately, the argument from commitment is defeated.

In the example dialogue above, it is pretty clear from the evidence of hiswords and actions that Ed is committed to communism, so the first criticalquestion can easily be answered. But Ed could possibly reply, using thesecond critical question, by explaining that in this particular case, hethinks the union is wrong, because their demands are excessive and willbankrupt the company, putting everyone out of work. So it is possibleto reply to an argument from commitment by saying that normally youwould be committed to a particular policy, but that this particular case isan exceptional one for you. There is the possibility of an exception to therule. The bringing forward of such evidence in asking the second criticalquestion would defeat the argument.

The second critical question has to do with what the respondent iscommitted to in the premise, in relation to what he is committed to in theconclusion. In some cases, the same commitment can be involved both inthe premise and in the conclusion. For example, consider the followingdialogue.

Bob: Ed, you are a communist, aren’t you?

Ed: Of course. You know I am passionate about that.

Bob: Well, then I assume you advocate the communist position, taking theunion side in the recent labor dispute.

In this example dialogue, the proposition A, Ed’s being a communist, isidentical or at least very close to being identical to B, in both the premiseand the conclusion of the inference. By contrast, in the previous exampledialogue, there is much more of a difference in the relationship between



the two propositions. In examining any case, a careful look at the actualwording of the two propositions is helpful.

One way that argument from commitment can be abused is throughthe committing of the straw man fallacy of distorting or exaggerating anopponent’s position, in order to make it easier to attack and refute it asimplausible. Environmental debates provide classic cases of such argu-ments, as the following example will indicate.

Stewart and Margo are having a dispute on environmental issues,where Margo has taken a moderate position to the effect thatdevelopment should be allowed only if it is sustainable and effortsare made not to pollute the environment. Stewart replies, “I seethat you are one of those extreme protectionists who think thatthe earth should be a pristine wilderness where all industrialdevelopment is forbidden.”

Much here depends on the prior context of the dialogue, and the partic-ular propositions that Margo has committed herself to accepting as herposition. But suppose her commitments are not as extreme as the rad-ical view portrayed by Stewart and that her form of environmentalismis moderate. In such a case, her point of view has been misrepresentedby him and made to appear more extreme than it really is. This kind ofargumentation is merely a tactic to make it easier to refute her argument.As shown by this case, the second critical question is extremely impor-tant. The proposition that the proponent is committed to might not be thesame as the proposition attributed to her by the respondent. If there is adifference between the two propositions, there has to be a reason to drawthe inference from the one to the other. Thus the use of argument fromcommitment is sometimes dangerous. It can even be used as a misleadingand erroneous move to attack an opponent in a dialogue.

Argument from commitment can be used in an even stronger way toinfer that a respondent is inconsistent in his commitments. Suppose thatin another case, Ed had actually taken the side of the management in arecent labor dispute. In such a case, Bob might use argument from com-mitment in a negative fashion to draw the conclusion that Ed’s conductimplies an inconsistency. Consider the following dialogue.

Bob: You are a communist, aren’t you?

Ed: Of course. You know that I have often said so.

Bob: Well, you say you are a communist, but you were against the unionside in the recent labor dispute, showing that you are not a communist.



In this example dialogue, Bob is drawing the conclusion that Ed’s com-mitments are inconsistent. Or at least he is arguing that they appear tobe inconsistent and that the apparent inconsistency raises doubts aboutwhether Ed’s professed commitments are really his commitments afterall. Bob says, “Well, you say you are a communist,” suggesting that Edmay not really be a communist at all, because “actions speak louder thanwords” when it comes to revealing one’s commitments. This kind of attackcan be used to make an arguer appear to be illogical or even hypocritical,as is shown in the analysis of various kinds of personal attack argumentsin section 9 below.

To reply to this kind of negative use of argument from commitment, asin his response to the third critical question, above, Ed has to offer somekind of account giving his reasons why he did what he did, showing howthis case was exceptional. Ed has to go further into the details of the case torestore the consistency of his commitments. This negative use of argumentfrom commitment, called argument from inconsistent commitment, hasthe following general form.

ARGUMENTATION SCHEME FOR ARGUMENT FROM INCONSISTENT

COMMITMENT

INITIAL COMMITMENT PREMISE: a has claimed or indicated that he iscommitted to proposition A (generally or in virtue of what he said inthe past).

OPPOSED COMMITMENT PREMISE: Other evidence in this particularcase shows that a is not really committed to A.

CONCLUSION: a’s commitments are inconsistent.

To reply to the use of an argument from inconsistent commitments, arespondent may need to go further into the details of the case or the natureof his commitment to show why the inconsistency is only apparent andnot real. Alternatively, if he admits the inconsistency is real, he mustsomehow explain how the conflict arose.

The critical questions matching the argument from inconsistent com-mitments are the following.

1. What is the evidence supposedly showing that a is committed to A?2. What further evidence in the case is alleged to show that a is not

committed to A?



3. How does the evidence from 1 and 2 prove that there is a conflict ofcommitments?

The problem generally with arguments from inconsistent commitmentsis that an action attributed to someone might suggest that they are notcommitted to a policy, but how an action is interpreted in a particularsituation, as expressing a commitment, is often highly subject to dispute.If I am a professed vegetarian but eat beef on one occasion, does thataction mean that I am now committed to a policy of eating beef? Maybenot, because I may just have slipped up on one occasion or been veryhungry when no vegetarian food was available. What an action implies,as a commitment, may not be so easy to judge and may require looking ata body of evidence in a given case.

EXERCISE3.7

Analyze the arguments in the following dialogues by identifying the argu-mentation scheme of the argument put forward by one party. Identify thepremises and conclusion of the argument. If there are any questionableaspects of the argument that should be considered, identify critical ques-tions that should be asked.

Dialogue (a)

RON: Rose, you are a Catholic, aren’t you?

ROSE: Yes, you know that.

RON: Well, then you must be voting against the Democrats in the next elec-tion, because they support abortion on demand.

Dialogue (b)

TINA: You believe in the theory of evolution, don’t you, Tom?

TOM: Well yes, I suppose I do accept that theory.

TINA: So you are one of those godless materialists who rejects equal rightson the grounds that all life should be a struggle where only the fittest shouldsurvive.

Dialogue (c)

BRUTUS: You said you were a supporter of free enterprise, right, Barbara?

BARBARA: That’s what I maintained, yes.

BRUTUS: But last week you said you approved of the new policy of tight-ening controls over emission of industrial pollution in the Bay Area.So you are not really committed to private enterprise at all, in thatinstance.



Dialogue (d)

SENATOR S: Back ten years ago, you condemned our policy on the groundsit contributed to inflation.

SENATOR T: Well yes, I did.

SENATOR S: But while your party has been in power, there has been moreinflation than ever.

EIGHT Ad Hominem Arguments

In any of the kinds of conversational frameworks in which people reasonwith each other, despite the opposition and partisanship characteristic ofmany kinds of dialogue, there must also be a presumption that in orderto achieve collaborative goals, participants must observe rules of politeconversation. Arguers must be able to trust each other, to some extent atleast, to be informative and relevant, to take turns politely, and to expresstheir commitments clearly and honestly. Without this kind of collabora-tion in contributing to a dialogue, argument, of a kind that uses reasoningto fulfill its goals of dialogue interaction, would not be possible. For thesereasons, attacking the other party’s honesty or sincerity in argument is apowerful move. Such an argument leads one to the conclusion that sucha person lacks credibility as an arguer who can be trusted to play by therules. This argument is so powerful because it suggests that such a personcannot ever be trusted and that therefore whichever argument they use, itmay simply be discounted as worthless. Thus the person attacked cannotmeaningfully take part in the dialogue any longer, no matter how manygood arguments they seem to have. Because they are so powerful anddangerous, ad hominem arguments have often been treated in the past asfallacious. Their use in negative campaign tactics in political argumenta-tion is notorious. But they can sometimes be reasonable arguments. Forexample, in legal argumentation in a trial, it can be legitimate for a cross-examining attorney to question the ethical character of a witness. Thelawyer may even argue that the witness has lied in the past and use thisargument to raise questions about his character for honesty. But before wecan evaluate such ad hominem arguments, it is necessary to know whatform they take.

The simplest form of the ad hominem, or personal attack, argument isthe direct or personal type, often called the abusive ad hominem argumentin logic textbooks.


8. Ad Hominem Arguments 123

ARGUMENTATION SCHEME FOR THE DIRECT AD HOMINEM ARGUMENT

CHARACTER ATTACK PREMISE: a is a person of bad character.

CONCLUSION: a’s argument should not be accepted.

In this type of argument, a is the proponent of an argument that has beenput forward. The premise that is alleged is that a is a person of bad char-acter. What is normally cited is some aspect of a’s character as a person,and often, character for veracity is the focus of the attack. For example,the allegation may be, “He is a liar!” The attack is directed to destroyingthe person’s credibility, so that his argument is discounted or reduced inplausibility because of the reduction in credibility of the arguer. Thus thistype of attack is particularly effective where a person’s argument dependson his presumed honesty or good character for its plausibility.

The critical questions appropriate for the direct ad hominem argu-ment are the following.

1. How well supported by evidence is the allegation made in the char-acter attack premise?

2. Is the issue of character relevant in the type of dialogue in which theargument was used?

3. Is the conclusion of the argument that A should be (absolutely)rejected, even if other evidence to support A has been presented, or isthe conclusion merely (the relative claim) that a should be assigned areduced weight of credibility as a supporter of A, relative to the totalbody of evidence available?

How these critical questions work can be illustrated by the followingcase.

Paul Johnson, in his book Intellectuals19 wrote biographicalaccounts of several famous intellectuals showing the persons inquestion to be disorderly and unethical in their private lives. OfKarl Marx, Johnson wrote that Marx was lazy in collecting facts,and often dishonest in reporting them (pp. 68–71), and thereforehe could not be trusted to use factual evidence in an objectiveway (p. 69).

19 London: Weidenfeld and Nicholson, 1988.



To evaluate this use of the direct ad hominem argument, the three crit-ical questions need to be asked. First, was it true that Marx was lazy incollecting facts and often dishonest in reporting them? To answer thisquestion, we have to look at the cases in point cited by Johnson and askwhether the incidents he cites show that Marx was lazy or dishonest. Toanswer the second critical question, we have to ask what the purpose ofJohnson’s book is. The purpose of the book is to attack the credibility ofseveral intellectuals, and intellectuals generally, by showing that they hadbad character, as shown by their inability to run their own personal livesin an ethical way. Johnson’s attack on Marx is relevant, then, in the sensethat it contributes to this purpose. (A later chapter examines in greaterdepth the question of what relevance is generally.)

The third critical question can be put as follows. Is the conclusion ofJohnson’s argument that Marx’s theory of economics should be absolutelyrejected, or is Johnson’s conclusion merely that Marx should be assigneda reduced weight of credibility, as a supporter of his theory, relative tothe total body of evidence available? It would seem that the latter is thebetter interpretation of Johnson’s argument. He is claiming not that Marx’stheory is false or has been absolutely refuted as an economic theory byhis ad hominem argument. He is claiming only that Marx is not a crediblesupporter of his theory. But even this weaker form of argument representsa serious attack on Marx’s advocacy of his theory.

The weakest aspect of Johnson’s ad hominem argument against Marxis on the count of the first question. But generally, his argument used toquestion Marx’s credibility carries weight as a plausible argument, pro-vided support can be given to Johnson’s claim that Marx was lazy anddishonest in specific instances. If a person has a bad character in certainrespects, it is perfectly legitimate for a biographer to argue that he had abad character, by citing facts to support her thesis.

In many textbooks, the direct ad hominem argument is called‘abusive’, suggesting that it is a fallacious argument and is always wrong.However, as the example above shows, sometimes direct ad hominemarguments can be reasonable, if they are based on facts that support themand if character is relevant as an issue in the dialogue. Still, it is true thatthe direct ad hominem argument is little more than ‘mud slinging’ or useof personal attack to try to discredit someone, often merely by innuendoor suggestion, with no facts presented to back up the allegation. Thus,care is needed in dealing with ad hominem arguments.

The circumstantial ad hominem argument is a variant on the directone based on argument from inconsistent commitment. The allegation of



inconsistent commitment is used to suggest that the arguer is not sin-cere in following the conclusion of his own argument. As in the follow-ing classic case of the circumstantial ad hominem argument, the alle-gation can be expressed by the saying, “You don’t practice what youpreach.”

Parent: There is strong evidence of a link between smoking and chronicobstructive lung disease. Smoking is also associated with many other seriousdisorders. Smoking is unhealthy. So you should not smoke.

Child: But you smoke yourself. So much for your argument against smoking!

In this dialogue on smoking, the child’s argument against the parent is anargument from inconsistent commitment, but it is also a circumstantialad hominem argument. Indeed, the circumstantial ad hominem argumentis a special subtype of argument from inconsistent commitment that hasthe following form.

ARGUMENTATION SCHEME FOR THE CIRCUMSTANTIAL AD HOMINEM

ARGUMENT

ARGUMENT PREMISE: a advocates argument α, which has propositionA as its conclusion.

INCONSISTENT COMMITMENT PREMISE: a is personally committed tothe opposite (negation) of A, as shown by commitments expressed inhis or her personal actions or personal circumstances expressing suchcommitments.

CREDIBILITY QUESTIONING PREMISE: a’s credibility as a sincere personwho believes in his own argument has been put into question (by thetwo premises above).

CONCLUSION: The plausibility of a’s argument α is decreased ordestroyed.

The circumstantial ad hominem argument is a chain of argumentationbased on combining argument from inconsistent commitment with thedirect ad hominem argument. The first two premises represent an argu-ment from inconsistent commitment. Together they lead to the thirdpremise, the credibility questioning premise, which is the conclusion ofthe argument from inconsistent commitment. But this conclusion ques-tions the arguer’s character as a sincere person. Thus it functions, in turn,as a premise that leads to the final conclusion of the circumstantial adhominem argument above. If the arguer a is a person of bad character,



and his argument depends on his good character (because it dependson his credibility), then the plausibility of his argument is weakened ordestroyed by the inconsistent commitment found in it.

The thrust of the child’s argument in the smoking case is based onthe perception of the parent’s inconsistent commitments, as expressedin combining the argument premise and the inconsistent commitmentpremise. The parent advocates nonsmoking, but at the same time, sheherself smokes. This combination of premises leads to the credibilityquestioning premise. If the parent is inconsistent in this way, it plausi-bly follows that the parent’s belief in her own argument is open to doubt.That is, the parent lacks credibility in the child’s eyes. Therefore, the childreasons, the parent’s argument can be rejected.

There are many critical questions appropriate for the circumstantialtype of ad hominem argument, but in this case, the following four arethe most important ones to consider. Credibility of a participant in dia-logue is an important notion in the third premise. An arguer’s credibilityis enhanced by showing that she has good character of a kind that makesher a cooperative and trustworthy participant who helps a dialogue. Con-versely, an arguer’s credibility can be attacked or destroyed by showingthat she has bad character – for example, if she has lied or if she is notsincere in believing what she says. Thus, credibility can enhance the plau-sibility of an argument or detract from it.

1. Is there a pair of commitments that can be identified, shown by evi-dence to be commitments of a, and taken to show that a is practicallyinconsistent?

2. Once the practical inconsistency is identified that is the focus of theattack, could it be resolved or explained by further dialogue, thus pre-serving the consistency of the arguer’s commitments in the dialogueor showing that a’s inconsistent commitment does not support theclaim that a lacks credibility?

3. Is character an issue in the dialogue, and more specifically, does a’sargument depend on his or her credibility?

4. Is the conclusion the weaker claim that a’s credibility is open to ques-tion or the stronger claim that the conclusion of α is false?

Now let’s see how the scheme applies to the smoking example. Withrespect to the first critical question, the practical inconsistency is easilyidentified: The parent claims you should not smoke (as a general prac-tice), but the parent smokes herself. The second critical question is howserious the inconsistency is. In this case, the parent could reply, “Yes, Ismoke, but I am trying my best to give it up. Once you start, it is hard to



stop.” If the dialogue were to be extended in this way, the parent couldtake some of the sting out of the claim of inconsistency. The third ques-tion is whether, as a result of the above considerations, the parent lackscredibility. The answer is, to some extent, yes. But some credibility canbe restored by the kind of explanation cited above.

In this case, the most important critical question is the fourth one.The problem with the child’s reaction is that he appears to reject the par-ent’s conclusion (that smoking is unhealthy) outright, declaring, “So muchfor your argument against smoking!” This response is an overreaction,because it could quite possibly be that the parent has presented good evi-dence to support this conclusion. By throwing the whole argument outthe window, the child could be making a big mistake. But if the child’s adhominem argument is only the weaker claim that the parent’s credibilityis open to question, on grounds of her being a smoker, then his standpointcould be quite a reasonable one.

EXERCISE3.8


(a) Bob Smith was once accused of sexual harassment. So it would bea mistake to take the views of this reprehensible pervert seriously.Therefore his theory about the disappearance of dinosaurs from theearth should be rejected.

(b) Child to Parent: Your argument that I should stop stealing candyfrom the corner store is no good. You told me yourself just a weekago that you too stole candy when you were a kid.

(c) Senator X: High taxes and too many regulations on business arethe worst things for the economy, and these things are not underreasonable control at all.Senator Y: When your party was in power, you drove taxes up torecord levels, and introduced many new regulations on business.So you are either illogical or, more likely, a hypocrite who doesn’tbelieve a word he says.

(d) A sign on a bumper sticker reads, “What’s more ridiculous than ananti-hunter who eats meat?”

(e) Mr. S. denounced investors who take advantage of tax loopholesin the law, arguing that these bad practices ruin the economy forthe average person. Mr. T. replied that S. himself had recentlytaken advantage of a “quick-flip” tax reduction scheme that wastechnically legal but was really a clever device for tax avoidance.



Mr. S. replied that you shouldn’t condemn anyone for taking advan-tage of the law and that what he did was legal. So he, like any citizen,should be able to take advantage of legitimate deductions.

(f) A politician in office charged that government policy is undulyinfluenced by corporations that give money to the party. Criticspointed out that private companies had donated millions of dollarsin contracts and grants to her own leadership campaign. Citing alist of such donations she recently received, they argued that hercondemnation of business donations contradicted her own solici-tation and acceptance of such funding. Her spokesperson repliedthat she was only speaking philosophically about an ideal worldwhen she called for a political party system funded only by publicmoney.

NINE Argument from Verbal Classification

Argument from verbal classification concludes that a particular thing hasa certain property on the grounds that this thing can be classified undera general category of things that have this property. A simple example isthe following argument.

All dolphins are classified as mammals.

Flipper is a dolphin.

Therefore, Flipper is a mammal.

In this case, the classification of dolphins as mammals is determined bythe science of biology. To the extent that the classification of all dolphins asmammals is not subject to exceptions or to borderline cases, the inferencein this example may be classified as deductively valid.

But arguments from verbal classification often rest on classificationsbased on word usage in everyday (non-scientific) speech. Consider thefollowing example.

Anyone with net assets of over two billion dollars is wealthy.

Sarah has net assets of over two billion dollars.

Therefore, Sarah is wealthy.

The term ‘wealthy’ is vague in ordinary usage. But it is uncontroversial toclassify anyone with net assets of over two billion dollars as wealthy. Forsuch a total of net assets is beyond the gray area where there would bedisputes about whether someone would rightly be classified as wealthy


9. Argument from Verbal Classification 129

or not. Thus despite the vagueness of the word ‘wealthy,’ the argumentfrom verbal classification in the example above can also be evaluated asbeing deductively valid.

However, some cases of classifications are more subject to doubt.Suppose the first premise in the example just above was the statement,“Anyone with net assets of over a million dollars is wealthy.” Whetherthis statement is acceptable depends on the context of its use. In somecountries, it would be true to say that anyone with net assets of over amillion dollars is “wealthy.” But in Saudi Arabia, for example, this clas-sification would not be accepted as plausible. It not only would leaveroom for doubt, it would be rejected as a correct account of the way theword ‘wealthy’ is used.

Classification may be set in place by conventions of scientific termi-nology or by common usage of terms in everyday discourse. But in somecases, it can also be set in place by legal definitions of terms such as‘murder’ or ‘capital gains.’ In any event, although some terms on whichclassifications are built are clearly defined in a way that is well estab-lished, other terms are highly subject to disputation. In these latter casesespecially, it is important to raise critical questions about arguments basedon a verbal classification.

The argument from verbal classification has the following generalform, where a is an individual entity, x is a variable ranging over suchentities, and F is a property.

ARGUMENTATION SCHEME FOR ARGUMENT FROM VERBAL

CLASSIFICATION

INDIVIDUAL PREMISE: a has property F.

CLASSIFICATION PREMISE: For all x, if x has property F, then x can beclassified as having property G.

CONCLUSION: a has property G.

The critical questions appropriate for the argument from verbal classifi-cation are the following.

1. What evidence is there that a definitely has property F, as opposedto evidence indicating room for doubt on whether it should be soclassified?

2. Is the verbal classification in the classification premise based merelyon a stipulative or biased definition that is subject to doubt?



A stipulative definition is one that is invented, as opposed to repre-senting an established or widely accepting meaning. For example, theterm ‘quark’ was introduced in physics to name a special type of sub-atomic particle. Or in economics, a recession may be stipulatively definedas two quarters of negative growth in GNP. This stipulative definitionmay not agree completely with the way the term is used in everydayspeech, but it could arguably be put forward to represent a technicaluse of the term in economics as a discipline. These stipulative defini-tions seem reasonable, in context, but in some cases such definitions aremore questionable because they have a value aspect, a kind of bias orspin that makes one’s viewpoint look good or an opposed viewpoint lookbad. Such use of emotionally loaded terms often needs to be criticallyquestioned.

An example showing the importance of critically questioning argu-ment from a verbal classification could occur in the case of the dialogueon tipping where Helen might use the following argument to support herviewpoint.

Tipping is elitist.

Therefore, tipping should be discontinued.

This argument from verbal classification is quite dangerous for Bob’s sideof the dispute. If Bob accepts the premise that tipping can be classifiedas an elitist practice, then by argument from verbal classification, Helen’sthesis that tipping should be discontinued directly follows by a struc-turally correct inference. Unless Bob critically questions this use of argu-ment from verbal classification, he loses the dispute. Bob has two waysto go. He can directly challenge Helen’s premise that tipping is elitist, byasking the first critical question. Or he can pose the second critical ques-tion by arguing that even if tipping is elitist to some degree or in certainrespects, this kind or degree of elitism is not necessarily a bad thing. Inother words, he can question whether a practice ought to be discontinuedjust because it can be classified as elitist.

Arguments from verbal classification frequently lead to verbal dis-putes about the real meaning of a term. But they can be quite hard todefend against if the term in question has strong negative connotationsof some sort in everyday usage. Once a stigma of this kind is affixed to athesis, it has a staining effect that is difficult to remove. In the dialogueabove, for example, Bob may try to turn the tables by calling Helen’s the-sis communistic or using some other term to describe it that is generally


9. Argument from Verbal Classification 131

perceived to be pejorative. In some cases, then, it is not hard to see howargument from verbal classification can lead into name calling and adhominem attacks.

Another important thing about argument from verbal classification isthat this type of argument is frequently used in an aggressive way to bringpressure to bear against an opponent by the use of an argument from averbal classification that is biased to one side of a dialogue. Suppose thattwo people are having a theological dialogue on the issue of whether Godis a trinity or a unity, and the proponent of the trinity thesis argues againstthe unity defender using the following argument.

Your thesis is heresy.

Therefore, your thesis is wrong.

This use of argument from a verbal classification seems to have the unitydefender on the ropes, assuming (as seems plausible perhaps to the dis-putants) that anything that can be classified as heresy sounds pretty badand must have something wrong with it. This implication could stem fromthe negative connotations of the term ‘heresy’ – it sounds like somethingbad. But it should be questioned, by asking the first critical question above,whether the thesis in question really can be classified as heresy. And evenif it can, it should be questioned, by asking the second critical questionabove, whether all heresy is really all that bad. After all, ‘heresy’ seems tobe stipulatively defined as any view that is against church dogma. Whatneeds to be recognized, then, is that ‘heresy’ is a kind of biased term, asused in the theological dialogue above. For any view that departs fromthe trinity arguer’s view in the dialogue above can be classified as heresy.Heresy is used not only as a derogatory term, or at least it is meant to be, bythe proponent of the argument in the dialogue above. Any view opposedto his own (the official church view) is thereby automatically classifiedas heresy, permitting no opposition to this view in the dialogue. Whata respondent needs to do is to challenge the classification of his thesisunder this derogatory term by questioning the biased use of this stipulativeterm.

Disputes about arguments from verbal classification are sometimesportrayed as trivial. But it is important to recognize that such argumentscan be extremely powerful and significant, because their consequencescan be enormously important. Such arguments can also be disputed atgreat length and, in some instances, at great cost in the courts. Considerthe following example of this sort.



A private corporation wants to build a new housing developmentin area X. However, there is a law that if any area is classifiedas a wetland, no commercial development of any sort can takeplace on it. Any area that has wildfowl on it and that has overtwo hundred square meters of water surface during the month ofJuly is classified as a wetland. An environmental group claimsthat area X meets these criteria and is therefore a wetland.

In this example, the environmental group uses the argument from verbalclassification to make their case that area X is a wetland. If they win thiscase in court, the corporation will not be able to go ahead with their plansfor developing area X.

EXERCISE3.9


(a) All kangaroos are marsupials. Jumper is a kangaroo. Therefore,Jumper is a marsupial.

(b) Anyone who believes in hiring people strictly on merit is taking anelitist viewpoint. Ben believes in hiring people strictly on merit.Therefore, Ben is an elitist.

(c) At issue in a trial was whether a man who had driven a bicyclewhile intoxicated had violated the law against drunk driving. Thekey question was whether a bicycle could be considered a vehicle.The court decided in the negative.

(d) Your argument supports free trade. Therefore, it is a globalist viewthat supports the big corporations that are against human rights.

TEN Summary

Below are listed all the argumentation schemes covered in this chapter,with the set of critical questions corresponding to each form of argument.

ARGUMENTATION SCHEME FOR ARGUMENT FROM POSITION TO KNOW

POSITION TO KNOW PREMISE: a is in a position to know whether A is trueor false.

ASSERTION PREMISE: a asserts that A is true (false).

CONCLUSION: A is true (false).


10. Summary 133

1. Is a in a position to know whether A is true (false)?

2. Is a an honest (trustworthy, reliable) source?

3. Did a assert that A is true (false)?

ARGUMENTATION SCHEME FOR APPEAL TO EXPERT OPINION

MAJOR PREMISE: Source E is an expert in subject domain S containingproposition A.

MINOR PREMISE: E asserts that proposition A (in domain D) is true (false).


1. Expertise Question: How credible is E as an expert source?

2. Field Question: Is E an expert in the field that A is in?

3. Opinion Question: What did E assert that implies A?

4. Trustworthiness Question: Is E personally reliable as a source?

5. Consistency Question: Is A consistent with what other experts assert?

6. Backup Evidence Question: Is E’s assertion based on evidence?

ARGUMENTATION SCHEME FOR APPEAL TO POPULAR OPINION

GENERAL ACCEPTANCE PREMISE: A is generally accepted as true.

PRESUMPTION PREMISE: If A is generally accepted as true, there exists apresumption in favor of A.

CONCLUSION: There exists a presumption in favor of A.

1. What evidence, such as a poll or an appeal to common knowledge,supports the claim that A is generally accepted as true?

2. Even if A is generally accepted as true, are there any reasons fordoubting it is true?

ARGUMENTATION SCHEME FOR ARGUMENT FROM ANALOGY

SIMILARITY PREMISE: Generally, case C1 is similar to case C2.

BASE PREMISE: A is true (false) in case C1.

CONCLUSION: A is true (false) in case C2.

1. Are there differences between C1 and C2 that would tend to under-mine the force of the similarity cited?

2. Is A true (false) in C1?

3. Is there some other case C3 that is also similar to C1, but in which Ais false (true)?



ARGUMENTATION SCHEME FOR ARGUMENT FROM CORRELATION TO CAUSE

CORRELATION PREMISE: There is a positive correlation between A and B.

CONCLUSION: A causes B.

1. Is there really a correlation between A and B?

2. Is there any reason to think that the correlation is any more than acoincidence?

3. Could there be some third factor, C, that is causing both A and B?

ARGUMENTATION SCHEME FOR ARGUMENT FROM POSITIVE CONSEQUENCES

PREMISE: If A is brought about, good consequences will plausibly occur.

CONCLUSION: A should be brought about.

ARGUMENTATION SCHEME FOR ARGUMENT FROM NEGATIVE CONSEQUENCES

PREMISE: If A is brought about, bad consequences will plausibly occur.

CONCLUSION: A should not be brought about.The following three critical questions match the argumentation schemesfrom either positive or negative consequences.

1. How strong is the probability or plausibility that these cited conse-quences will (may, might, must) occur?

2. What evidence, if any, supported the claim that these consequenceswill (may, might, must) occur if A is brought about?

3. Are there consequences of the opposite value that ought to be takeninto account?

ARGUMENTATION SCHEME FOR THE SLIPPERY SLOPE ARGUMENT

FIRST STEP PREMISE: A0 is up for consideration as a proposal that seemsinitially like something that should be brought about.

RECURSIVE PREMISE: Bringing up A0 would plausibly lead (in the givencircumstances, as far as we know) to A1, which would in turn plausiblylead to A2, and so forth, through the sequence A2, . . . , An.

BAD OUTCOME PREMISE: An is a horrible (disastrous, bad) outcome.

CONCLUSION: A0 should not be brought about.

1. What intervening propositions in the sequence linking up A0 withAn are actually given?

2. What other steps are required to fill in the sequence of events to makeit plausible?


10. Summary 135

3. What are the weakest links in the sequence, where specific criticalquestions should be asked about whether one event will really lead toanother?

ARGUMENTATION SCHEME FOR ARGUMENT FROM SIGN

SPECIFIC PREMISE: A (a finding) is true in this situation.

GENERAL PREMISE: B is generally indicated as true when its sign, A, istrue.

CONCLUSION: B is true in this situation.

1. What is the strength of the correlation of the sign with the eventsignified?

2. Are there other events that would more reliably account for the sign?

ARGUMENTATION SCHEME FOR ARGUMENT FROM COMMITMENT

COMMITMENT EVIDENCE PREMISE: In this case it was shown that a is com-mitted to proposition A, according to the evidence of what he said ordid.

LINKAGE OF COMMITMENTS PREMISE: Generally, when an arguer is com-mitted to A, it can be inferred that he is also committed to B.

CONCLUSION: In this case, a is committed to B.

1. What evidence in the case supports the claim that a is committed toA, and does it include contrary evidence, indicating that a might notbe committed to A?

2. Is there room for questioning whether there is an exception in thiscase to the general rule that commitment to A implies commitmentto B?

ARGUMENTATION SCHEME FOR ARGUMENT FROM INCONSISTENT

COMMITMENT

INITIAL COMMITMENT PREMISE: a has claimed or indicated that he is com-mitted to proposition A (generally or in virtue of what he said in the past).

OPPOSED COMMITMENT PREMISE: Other evidence in this particular caseshows that a is not really committed to A.

CONCLUSION: a’s commitments are inconsistent.

1. What is the evidence supposedly showing that a is committed to A?

2. What further evidence in the case is alleged to show that a is notcommitted to A?



3. How does the evidence from 1 and 2 prove that there is a conflict ofcommitments?

ARGUMENTATION SCHEME FOR THE DIRECT AD HOMINEM ARGUMENT

CHARACTER ATTACK PREMISE: a is a person of bad character.

CONCLUSION: a’s argument should not be accepted.

1. How well supported by evidence is the allegation made in the char-acter attack premise?

2. Is the issue of character relevant in the type of dialogue in which theargument was used?

3. Is the conclusion of the argument that A should be (absolutely)rejected, even if other evidence to support A has been presented, oris the conclusion merely (the relative claim) that a should be assigneda reduced weight of credibility as a supporter of A, relative to the totalbody of evidence available?

ARGUMENTATION SCHEME FOR THE CIRCUMSTANTIAL AD HOMINEM

ARGUMENT

ARGUMENT PREMISE: a advocates argument α, which has proposition Aas its conclusion.

INCONSISTENT COMMITMENT PREMISE: a is personally committed to theopposite (negation) of A, as shown by commitments expressed in his orher personal actions or personal circumstances expressing such commit-ments.

CREDIBILITY QUESTIONING PREMISE: a’s credibility as a sincere person whobelieves in his own argument has been put into question (by the twopremises above).

CONCLUSION: The plausibility of a’s argument α is decreased ordestroyed.

1. Is there a pair of commitments that can be identified, shown by evi-dence to be commitments of a and taken to show that a is practicallyinconsistent?

2. Once the practical inconsistency is identified that is the focus ofthe attack, could it be resolved or explained by further dialogue thuspreserving the consistency of the arguer’s commitments in the dialogueor showing that a’s inconsistent commitment does not support theclaim that a lacks credibility?


10. Summary 137

3. Is character an issue in the dialogue, and more specifically, does a’sargument depend on his or her credibility?

4. Is the conclusion the weaker claim that a’s credibility is open toquestion or the stronger claim that the conclusion of α is false?

ARGUMENTATION SCHEME FOR ARGUMENT FROM VERBAL CLASSIFICATION

INDIVIDUAL PREMISE: a has property F.

CLASSIFICATION PREMISE: For all x, if x has property F, then x can beclassified as having property G.

CONCLUSION: a has property G.

1. What evidence is there that a definitely has property F, as opposedto evidence indicating room for doubt on whether it should be soclassified?

2. Is the verbal classification in the classification premise based merelyon a stipulative or biased definition that is subject to doubt?

argumentation schemes

Documents